Due to its openness and timeliness,the S&T Web information has become one of the most important resources for strategic intelligence monitoring.However,since S&T Web information is unstructured and lack of semantic description,it is a challenge to transfer the unstructured Web information into structured semantic knowledge.To solve this problem,the authors propose a method for structural monitoring of the S&T Web information resources.By using the knowledge extraction technologies,the authors firstly extract the knowledge objects as well as the relationship between objects from the Web resources and convert the free text into calculable structured knowledge units.Based on those extracted structured information,the authors build various kinds of monitoring models to realize research profiling for specific research fields.Based on those ideas,the authors implement the automated Web information monitoring system suitable for research field monitoring.A research profiling experiment also is carried out based on the semantic resources which are converted from the monitored Web data.
Zhixiong ZHANGXiaolin ZHANGJianhua LIUYimin ZOUJing XIELi QIANYing WANG
Intelligence attention model is a premise of implementing the structured monitoring of science and technology(S&T) information,and plays an important role in identifying intelligence value of S&T information and intelligence monitoring system.Analyzing the characteristics of scientific information based on monitoring the express in the energy technology domain,this paper has found that the object and its behavior have an important role in intelligence value judgment and have institutional particularity,and the number of important institutions in the domain has convergence.Using these characteristics,an intelligence attention model based on object behavior has been proposed,which not only focuses on the objects themselves,but also is more concerned about object behavior as well as the event behind the behavior.After evaluating and analyzing the application effect of this model,we find that this model can improve the degree of accuracy and refinement of intelligence value iudement.
